Sequential Injection Analysis
- Lesson 1: Introduction
- Lesson 2: Fundamentals of Flow Injection Analysis
- Lesson 3: Membrane Sampling Devices
- Lesson 4: Dispersion
- Lesson 5: Enrichment
- Lesson 6: Chemistry
- Lesson 7: Sequential Injection Analysis
- Lesson 8: Zone Fluidics
WHAT IS SIA?
SIA is an automated approach to sample handling that will allow you to automate manual wet chemistry procedures in a rapid, precise, and efficient manner. Small solution zones are manipulated under controlled dispersion conditions in narrow bore tubing.
Sharing many characteristics with FIA, some would argue that it is simply an extension of FIA. Nevertheless, more than 100 journal articles have been published on SIA since the first paper in 1990. While, like FIA, it is fundamentally dependent on the dispersion of zones in a flowing stream, conceptually, the practice of SIA is different from FIA.
Lets look at a simple FIA experiment and compare it to its SIA equivalent. Consider a single line FIA experiment where we inject a sample into a carrier stream containing a reagent. The FIA manifold is depicted in the above diagram. A sample is pumped into the sample loop of a two-position injection valve and the carrier is flowing constantly through the detector. The length of the sample loop determines the volume of sample injected. When the sample loop is loaded, the valve is switched and the sample is introduced into a flowing carrier stream. The carrier carries the sample through the reactor (usually a reaction coil) to the detector. En route, the sample reacts with the reagent to form a detectable species. The detectable species gives rise to a peak when it passes through the flow cell of the detector. A calibration curve is then used with the peak height, area, or width to determine the concentration of the analyte in the sample.
SIA on the other hand does not make use of an injection valve. Rather, a multi-position selection valve replaces the injection valve. Usually, the frequently used FIA peristaltic pump is replaced with a syringe pump and an additional coil called the holding coil is added. Compare the schematic of an SIA manifold to that of the FIA manifold. To achieve the same measurement as described above, the syringe is filled with carrier solution containing the reagent. Then the selection valve is advanced to a port that is connected to the sample line. A small volume of sample is drawn up into the holding coil. The flow program determines the volume of sample; viz. the volume of sample that is drawn up by the pump into the holding coil. The selection valve is then advanced to a port that is connected to the detector, and the carrier transports the sample through the reactor to the flow cell of the detector. Again, a detectable species is formed and is registered as a peak by the detector. The concentration of the analyte in the sample is determined in a similar manner as for FIA.
It is worth noting that at the moment of injection in an FIA experiment, an undispersed plug of sample is introduced into the carrier stream. In SIA, already during aspiration of the sample into the holding coil, dispersion begins to take place and the flow reversal that takes place when the sample is sent off to the detector plays a dramatic role in mixing the sample with the carrier. The next figure depicts graphically what happens at the point of sample injection (1) in an FIA manifold (top) and during the flow reversal inherent to SIA manifold (bottom) and a few seconds after the carrier has started to move (2) the sample towards the detector. This phenomenon may cause an SIA peak to look slightly different to an FIA peak. As long as there has been good mixing between sample and reagent, this will not affect quantification because samples and standards are treated alike. In fact, it has been shown that the flow reversal contributes significantly to the mixing of zones.
To further ensure good mixing of reagent and sample, we recommend the use of a knotted reactor such as our Super Serpentine Reactor.
More often than not, instead of including the reagent in the carrier, an SIA experiment is expanded so that the reagent is loaded as a separate zone. In this case the syringe is first filled with a simple carrier or buffer. After the sample zone has been drawn up into the holding coil, the selection valve is advanced to a port connected to a reagent reservoir and a small reagent zone is drawn up into the holding coil. In this way, it is possible to construct a stack of well defined zones which can be mixed together to give rise to a detectable species. You will appreciate that unlike FIA, which requires re-plumbing when a more complex chemical addition scheme is required, in SIA, all that is required is a change to the flow program. The manifold remains the same. The additional advantages of lowering reagent consumption and minimizing the production of potentially hazardous wastes are both important advantages of SIA.
The other ports of the selection valve can be used for calibration standards, additional reagents and as locations where more sophisticated operations such as dilution, trace enrichment, and incubation of reactants can take place. More about this later in the lesson.
A variation of SIA is sequential injection titration (SIT). In SIT, the reactor and detector in the above diagram are replaced with a stirred titration cell. Appropriate sensors, electrochemical or colorimetric, are placed in the titration cell which acts as flow cell as well as titration chamber. Sequential injection titrations are the subject of another lesson.
HOW DID SIA COME ABOUT?
SIA was developed by Ruzicka and Marshall at the University of Washington in response to an industry-initiated requirement for a more robust automated wet chemistry technique than FIA. At the Center for Process Analytical Science (CPAC), industry representatives challenged faculty and researchers to take FIA to the next level. Develop it to the point where its strengths were maintained and enhanced, but its limitations in the process environment were remedied. At other institutions, the problem was tackled by investigating the use of sensors. Other researchers explored miniaturization of FIA. Another group at CPAC looked at the development of a hybrid flow sensor called the Flow Probe.
In Ruzicka and Christian's group, the goal of simplifying the manifold, reducing reagent consumption, and relying more on micro processor controlled flow programming gave rise to SIA. Early workers in the field had to broaden their understanding of what makes for a good flow manifold. It soon became apparent that a key factor was going to be an understanding of factors that impact on zone penetration. Unlike FIA where the reagent zone is often merged with the carrier and therefore with the sample at a confluence point, in SIA interaction between the sample and the reagents takes place through zone penetration. In the following diagram, the zone of overlap represents the zone where chemical reactions will take place and the detectable species will form. The larger this zone of overlap, the more sensitive the measurement will be.
Since its description in 1990, researchers have exploited SIA for the analysis of compounds as diverse as radionuclides and several different bio-compounds. In many cases, its advantages when applied in the process environment have been exploited.
WHAT ARE THE ADVANTAGES AND DISADVANTAGES OF SIA?
While SIA makes use of a simpler flow manifold (this is particularly so for multi component chemistries), development of the SIA method is not as straight forward. Careful attention needs to be given to the design of the measurement sequence to ensure that adequate zone penetration has taken place.
Accurate measurement of sample and reagent zones necessitates microprocessor control. Of course, once the method has been developed, the microprocessor ensures slavish repetition of the optimized sequence.
Previous limitations associated with the use of syringe pumps (notably the need for a syringe fill cycle, and poor precision for sample volumes smaller than 10 ?l), no longer apply. Global FIA has developed a new pump called the milliGAT? that has all the advantages of a syringe pump but eliminates many of its limitations. In particular, the fill cycle and need to compromise syringe diameter in order to allow sufficient carrier volume for an experiment are conveniently overcome in the milliGAT??.
SIA has several advantages over FIA.
- Reagent use is drastically reduced. Typical FIA experiments make use of at least 1mL of reagent per measurement. SIA typically makes use of 50?l. This means that in a 24-hour period assuming one measurement per minute, the FIA analyzer would consume 1440ml of reagent. The SIA analyzer would consume 72ml. It has been noted that the most frequent reason for process analyzer failure is running out of reagents.
- Flow manifolds are simple and robust typically comprising a pump, selection valve, and detector connected by tubing. The same manifold can be used for widely different chemistries simply by changing the flow program rather than the plumbing. Analyzer maintenance is therefore simplified.
- The selection valve replaces the injection valve and provides a means for selecting different sample streams and calibrants. This enables convenient automated calibration.
- Components used in a SIA manifolds are amenable to laboratory, field, and plant operation. In addition to these, SIA enjoys all of the benefits of FIA.
WHAT EQUIPMENT IS NEEDED FOR SIA?
A typical SIA manifold has been described above and comprises the following main components:
- Selection valve
The following picture shows a four panel FloPro analyzer configured to do SIA.
Syringe pumps have been most widely used to aspirate zones and propel the stack of zones through the detector. Some researchers have used peristaltic pumps. The requirements for the pump are that it is precise, reproducible, bi-directional, and able to measure small volumes. Computer control is imperative. Cavro makes a reliable syringe pump and the Global FIA milliGAT? pump is expected to replace syringe pumps in future SIA instruments.
The selection valve must allow random access of the ports. Small dead volume and zero cross contamination between ports are essential features of a good selection valve. Valco valves have proved to be most suitable for SIA and these are available with between 6 and 28 ports. The 10 port multi-position valve is by far the most widely used.
|Reactor and Holding coil
Although the idea of knotting reactors in FIA has long been advocated, many users of these flow-based techniques make use of reaction coils. We have shown that if excellent mixing without increasing dispersion is required, then a serpentine flow path provides the optimum conditions. We have developed a Super Serpentine Reactor and have shown that it is superior to all present mixing reactors.
The wide range of detectors that are employed for FIA are suitable for SIA. The only requirement is that they be equipped with a flow cell. As for FIA, low dead-volume and immunity to bubbles are key requirements. We have a range of colorimetric and electrochemical detectors which have been developed for FIA and SIA.
The crux of SIA is the flow program. This sequence of events results in the assembly of the stack of zones in the holding coil and subsequent transport to the detector flow-cell. Microprocessor control is imperative. Several packages have been written to achieve this. We have a DOS-based package that has the functionality to provide for the requirements of a process analyzer. It is called Flow TEK?. A Windows package called FloStar(tm) is under development.
How would I develop an SIA method?
The development of an SIA method follows a similar procedure to the development of an FIA method. It would be worth reviewing the Chemistry tutorial at this point to refresh your memory on how method development is approached for FIA.
In general, the key steps in the development of a SIA method are as follows:
Identify a suitable chemistry of measurement
- A good starting point is to establish whether there are manual methods already in use for your samples. In addition to providing useful insights to a chemistry that has been proven, if there is an existing method, method validation is more easily achieved if the chemistry applied is common to both the manual and automated SIA method.
- Review the literature for methods to see whether there is any published literature on the analyte of interest. In this regard, the Flow Analysis Database (FAD) is a valuable resource. Most of the papers on SIA thus far have been devoted to the description of methods of analysis. Search for both FIA and SIA methods. If a paper is found, it will provide a useful starting point but you will probably have to optimize conditions for your particular sample.
- Refer to a good general reference on wet chemistry measurement chemistries. At the end of the Chemistry chapter in the Tutorial, there is a list of recommended references. Identify a suitable chemistry of measurement.
- If none of these resources provide a method, then the task becomes quite challenging and requires the development of the chemistry of measurement from first principles. That goes beyond the scope of this Tutorial.
Determine appropriate reagent concentrations and stochiometry
- A careful study of the method will soon reveal the reagents that are to be used and their relative quantities.
- The sequence of additions can be conveniently duplicated in an SIA experiment by aspirating appropriate zones of sample and reagent and building a stack of zones in the holding coil that will be mixed together as they are transported to the detector. Of course, in some instances, it may be necessary to pre-mix two reagents before exposing the mixture to the sample. It will now be apparent that this is easily achieved using SIA. Indeed this is one of the strengths and characteristics of SIA. Sample manipulation is controlled by flow programming or stated in another way, by manipulating the pump and selection valve under microprocessor control.
- Typical volumes used in SIA are on the ?l scale. It is not uncommon for the sample and reagent zones to be 20?l
- A useful means of establishing what the mixing patterns are is to sequentially introduce a suitable dye as each of the zones. You can simply use water or any non-detectable solution for the other zones. By overlaying the resultant dye profiles as recorded at the detector, you can see how each zone overlaps the others. Of course, this experiment does not take into account the affect of chemical reactions. It simply describes the physical zone penetration.
- Once the flow program has been established, you will have to test whether you have sufficient reagent to ensure that the method is not reagent starved. You do this by increasing the reagent concentration until further increase does not result in an increase of the signal. If it is reagent starved, you will either have to increase the reagent concentration, or the zone volume, or you could sandwich the sample between two zones of reagent. Of course, increasing the zone volume is only affective if there is adequate mixing.
Optimize all experimental parameters
- All parameters typically optimized in the development of the chemistry of a wet chemical procedure also should be optimized in the SIA methodology. These include the strength and pH of buffers, addition of masking agents, reaction and incubation times, and temperature. (Global FIA has a Reactor Heater that conveniently accepts metal-potted Super Serpentine reactors.)
- Physical parameters such as reagent volume, pump speed, and reactor length also affect the performance and robustness of the method.
Test the method using samples of known concentration
- Once the method has been developed, its analytical figures of merit can be determined. These include
- Precision of the method. Usually 10 replicates of a sample with concentration at mid range are measured and the %RSD (std deviation / mean expressed as a percentage)
- Long term and inter-laboratory precision. A good measure of the day to day a repeatability of the method and inter-laboratory repeatability provides a measure of the robustness of the method.
- Linear range. Standards spanning the useful range of the method are measured and a linear regression analysis is carried out. Goodness of fit as determined by the correlation coefficient r2 is frequently calculated and found to be better than 0.98.
- Detection limit. A common measure of detection limit is the concentration that corresponds to the signal equal to 3 times the standard deviation of the detector baseline.
- Determination limit. The determination limit is usually set at 10 times the standard deviation of the detector baseline.
- Measurement frequency. The number of measurements that can be carried out in an hour is quoted. When quoting this figure, any sample preparation prior to injection into the SIA manifold should be reported separately. This figure is derived from the time it takes the SIA instrument to do a single measurement.
- The accuracy of the method is tested by comparing the results obtained by an independent technique or by the analysis of certified reference materials.
- Explore whether commonly occurring matrix elements interfere with the measurement
What can I do with SIA?
A review of the present literature on SIA will give you a good feel for what is presently achievable. Some of the techniques that have been employed in sequential injection analyzers include:
- Trace enrichments
- Chemical masking
- pH adjustment
- Medium exchange miniature columns
- Phase exchange using membrane sampling devices
- Use of solid reagents
A wide range of detectors can be employed with SIA. These include
- UV, vis, and ir photometers
- Ion selective electrodes
- Potentiometric stripping analysis
- Other electrochemical techniques
This completes this session of our Web Tutorial.